How we do User Experience Research in agriculture
Balancing the particularities of agriculture with research methodologies is challenging, but necessary to deliver the best user experience.
Researching user behaviors and needs is essential to designing products that meet their expectations and provide a good experience. When the products are agricultural work tools, however, there are specific nuances that make the planning and execution of research more challenging.
It’s important to keep these in mind when planning research to gather data and make analyses that support good design decisions.
By getting hands-on, Syngenta Digital’s Product Design team has identified some best practices that make planning and conducting User Experience Research in agriculture more efficient.
Planning the research
In agriculture, there is a right time to plan, buy inputs, plant, manage the fields, harvest, and prepare the soil for the next season. The frequency and way these steps are carried out can vary widely depending on the crop and the region. Considering the agricultural schedule when planning research is necessary because growers tend to be unavailable during the busiest periods and inviting them to a 45-minute interview may sound inconvenient — as well as being ineffective, because they won’t agree to schedule it.
Besides, features need to be launched at the right time. A feature to track the harvest must be available to the user before it begins. We need to consider not only the grower’s time and the time taken to recruit, interview, analyze, and conclude the research, but also the time taken by Design and Engineering to iterate, test, and develop the feature. This is why user research works best when there is long-term planning based on the product roadmap.
Adapting the method to the user’s possibilities
Field research in agriculture tends to be expensive and time-consuming, due to the distances between farms and the office. It’s necessary to choose viable research methods — even if they’re not always ideal.
Online research tends to work but with a few caveats. It’s common to find growers with in-depth agronomic knowledge but limited tech-savviness. Difficulties opening prototypes or joining video calls can invalidate tests or delay interview schedules. In addition, being observed while browsing a product can be quite uncomfortable for these user profiles. All this needs to be considered when planning and analyzing the results.
Aligning with the team
In a complex market like agriculture, it’s harder to do research alone. The sooner we bring agronomists, product managers, and other experts into the process, the better. This way, we get valuable ideas and correct the course of the interview sooner, if necessary.
Aligning interview scripts with Syngenta agronomists and product managers always helps us to build the right questions.
In the Syngenta Digital Product Design team, this alignment is facilitated by Condens, a research repository tool that helps us gather interviews, organize findings, and make everything easily available to the entire team. This generates insights with previously mapped data and avoids rework.
Hands-on: Yield and revenue research case study
In 2023, the Design team dedicated to Cropwise Financials conducted research that perfectly illustrates how we put these guidelines into practice.
On Cropwise Financials our users track their revenue through their yield. They can then compare this to their input costs and understand their profitability at a field level. We needed to understand how to best build this feature to empower growers by enabling them to seamlessly track their yield sales and trace revenue back to the fields and crop zones.
The MVP version of the yield feature had already been released in early 2023, so our goal was to understand the needs of our users to build the feature out further.
Timeline
The release of this feature was set for mid to late August, aligning with the peak of the harvest season. To align with this delivery, we initiated the discovery work in late March/early April, allowing for iterative design and development.
Participant Recruitment
We worked with our product team as well as marketing resources to reach out to growers within the Agriedge program. Once interviews were scheduled, we created a participant record within Condens, allowing us to track our journey in speaking to specific users through multiple rounds of interviews.
Test Plan
Our research efforts were centered around three key goals:
Understanding User Expectations
We wanted to know what our users expected in a robust yield and revenue feature.
Recording Yield and Revenue
Understand how users are currently recording yield for their various crops.
Improving the Initial Version
Identifying areas for improvement in the MVP version of the yield feature, and understanding how to best align it with users’ needs.
Methods
We used the following methods during each interview session.
User Testing
We presented prototypes and early ideation concepts to users for their direct feedback. This was not usability testing as we were not evaluating whether a user could complete tasks with our design but rather if our design ideas accommodated their needs.
Contextual Inquiry
We also interviewed our participants to understand the journey of how they tracked yield and revenue. Our goal within the Design team is to build features that align with what users are doing so we always want to understand the deeper user journey outside of our products.
Analysis
Once the interviews were completed, we uploaded the video file of each to Condens. From there we were able to tag interviews and track common insights across each participant we spoke to.
Key Findings
After the research and analysis process, we developed the following insights:
Diverse Yield Tracking: Growers exhibited varied methods for tracking yield, dependent on the type of crop they were harvesting. The complexity arose from the fact that almost all growers cultivated multiple crop types, meaning that a single-form yield entry flow wouldn’t work for even one grower.
Load Ticket Customization: From dealing with a single load of harvest to managing numerous tiny loads for each crop, growers expressed the need for a customizable and reusable solution for tracking yield load tickets. We found that the way growers had yield data varied wildly and by crop. Some entered yield as a single load ticket at the end of the season, others had a large spreadsheet with hundreds of individual loads. Our solution needed to accommodate both.
Outcomes
Based on these key findings we worked to develop a central model that would allow users to track yield and revenue in a way that supported the diversity of possibilities. We landed on a template approach where users would be able to customize (or use pre-made default templates) the way they entered yield per crop. This would allow the user with multiple crop types to be able to collect their data and enter it to generate a single top-line revenue number for their organization.
The research we performed here was essential for us in defining our approach and because it was conducted early in the process, we reduced the need for costly adjustments later. The biggest value of this research was it aligned our Design and Product teams toward a single approach. This allowed us to move fast during the design and requirements process as we were then refining our approach rather than developing it. Then when we handed off designs to development, we were equipped with a total understanding of the pains and priorities of users, and consequently, we were well-backed to defend all the design decisions.
Having this deeper understanding of users behavior helped us to clarify the doubts of the development team so that we could all work towards delivering the same outcome, a feature that meets the real needs of the users, suited to their ways of working and respecting the multiplicity of agronomic management — essential requirements for providing a good user experience in agriculture.
— in collaboration with Nathalia Ilovatte
#syngenta #digital #reaserch #user #agriculture
